Publisher version

Downloaded from orbit.dtu.dk on: Dec 31, 2016
Boron-nitrogen based hydrides and reactive composites for hydrogen storage
Jepsen, Lars H.; Ley, Morten B.; Lee, Young-Su; Cho, Young Whan; Dornheim, Martin; Jensen, Jens
Oluf; Filinchuk, Yaroslav; Jørgensen, Jens Erik; Besenbacher, Flemming; Jensen, Torben René
Published in:
Materials Today
DOI:
10.1016/j.mattod.2014.02.015
Publication date:
2014
Document Version
Publisher final version (usually the publisher pdf)
Link to publication
Citation (APA):
Jepsen, L. H., Ley, M. B., Lee, Y-S., Cho, Y. W., Dornheim, M., Jensen, J. O., ... Jensen, T. R. (2014). Boronnitrogen based hydrides and reactive composites for hydrogen storage. Materials Today, 17(3), 129-135.
10.1016/j.mattod.2014.02.015
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal ?
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Materials Today Volume 17, Number 3 April 2014
RESEARCH: Review
RESEARCH
Boron–nitrogen based hydrides and
reactive composites for hydrogen storage
Lars H. Jepsen1, Morten B. Ley1, Young-Su Lee2, Young Whan Cho2,
Martin Dornheim3, Jens Oluf Jensen4, Yaroslav Filinchuk5,
Jens Erik Jørgensen6, Flemming Besenbacher7 and Torben R. Jensen1,*
1
Center for Materials Crystallography, Interdisciplinary Nanoscience Center and Department of Chemistry, Aarhus University, Langelandsgade 140,
DK-8000 Aarhus C, Denmark
2
High Temperature Energy Materials Research Center, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea
3
Helmholtz-Zentrum Geesthacht, Department of Nanotechnology, Max-Planck-Straße 1, 21502 Geesthacht, Germany
4
Department of Energy Conversion and Storage, Technical University of Denmark, Kemitorvet 207, DK-2800 Kgs. Lyngby, Denmark
5
Institute of Condensed Matter and Nanosciences, Université Catholique de Louvain, Place L. Pasteur 1, B-1348 Louvain-la-Neuve, Belgium
6
Department of Chemistry, University of Aarhus, 8000 Aarhus C, Denmark
7
Interdisciplinary Nanoscience Center (iNANO) and Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, DK-8000 Aarhus C, Denmark
Hydrogen forms chemical compounds with most other elements and forms a variety of different
chemical bonds. This fascinating chemistry of hydrogen has continuously provided new materials and
composites with new prospects for rational design and the tailoring of properties. This review highlights
a range of new boron and nitrogen based hydrides and illustrates how hydrogen release and uptake
properties can be improved.
Introduction
Water is the major naturally occurring liquid compound on earth
covering ca. 70% of the earth’s surface. Electrolysis powered by
renewable energy sources, for example, wind and solar energy
enables the splitting of water to H2 and O2. The stored energy can
be released as electricity and heat by reacting H2 and O2 to form
water in a fuel cell. The overall process is a closed sustainable
material cycle, where hydrogen is working as an energy carrier [1].
A remaining challenge is to store the significant amounts of
hydrogen [2–11].
Hydrogen bonds are vital for biological systems and life, the
unique properties of water, and so on. The dihydrogen bond is
defined as the interaction between a metal hydride bond (hydridic hydrogen) and an OH or NH group or other proton donor
(protic hydrogen). Unlike the classical hydrogen bond, the
dihydrogen bond can react in the solid state via elimination
of hydrogen by exchanging the weak Hd+ –dH interactions for
strong covalent bonds in H2, and thus may open new routes to
*Corresponding author:. Jensen, T.R. ([email protected])
the rational design of structures and hydrogen release reaction
mechanisms [12].
A rarer interaction is the hydrogen–hydrogen bond, which
occurs between two bonded hydrogen atoms with similar partial
charges, which may be significantly stronger than van der Waals
interactions and may play an important role for the physical
properties of solid molecular borohydrides, for example, Zr(BH4)4
[13]. The hydrogen molecule can also act as a ligand in some
complexes of the heavier d-block metals, for example, W or Re, and
the distance between the hydrogen atoms can be utilized to
distinguish between a M(h2-H2) complex with H–H bond length
in the range 0.74 to 1.38 Å and polyhydrido complexes with H–H
distances > 1.6 Å. The limiting value 0.74 Å is the internuclear
distance in the hydrogen molecule [14].
The ligands NH3 and BH4 are three dimensional, while NH2
and NH2 are two- and one-dimensional, respectively. The bond
length in BH4 is approx. 1.22–1.24 Å, while the bond length is
slightly shorter and decreasing from neutral NH3 (1.01 Å) toward
NH2 and NH2.
The hydrogen molecule, H2, has the lowest number of electrons
(2) of all molecules and therefore has the weakest physisorption
1369-7021/06/$ - see front matter ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.mattod.2014.02.015
129
RESEARCH
Materials Today Volume 17, Number 3 April 2014
RESEARCH: Review
interactions. Hydrogen adsorption is often observed to follow
Langmuir isotherms (i.e. monolayer adsorption), and the amount
of hydrogen adsorbed under saturation conditions is generally
proportional to the specific surface area determined by the BET
method for the porous adsorbent (known as Chahine’s rule) [15–
17]. A significant advantage is that the physisorption process has
fast adsorption/desorption kinetics.
Ionic, metallic and covalent bonds are also formed by hydrogen,
which has been discussed in another review in this issue of
Materials Today [18]. Altogether, this illustrates that hydrogen
forms a variety of different types of chemical bonds and interactions with matter and other elements and reacts with almost all
other elements in the periodic table [19]. This suggests that there is
still room for significant discoveries of a variety of novel hydrogen
containing materials. Furthermore, hydrogen can be released in a
chemical reaction between two or more hydrides denoted reactive
hydride composites (RHC) [8,11]. Dihydrogen bonding and the RHC
concept provide new schemes for the design and synthesis of new
materials with novel properties and for tailoring known materials
properties.
Reactive hydride composites
A promising approach for tailoring thermodynamic properties is
to allow two or more hydrogen-containing materials to react
during the release of hydrogen. Such hydride mixtures are denoted
Reactive Hydride Composites (RHC) [20]. In 2002, Chen et al. discovered reversible hydrogen release and uptake of lithium amide
hydride, LiNH2–LiH, according to reaction scheme (1) [21].
LiNH2 ðsÞ þ 2LiHðsÞ $ Li2 NHðsÞ
þ LiHðsÞ þ H2 ðgÞ $ Li3 NðsÞ þ 2H2 ðgÞ
(1)
The first step involves an amide–imide reaction, which may
involve hydrogen elimination via dihydrogen-bonded hydrogen,
but has also been suggested to involve ammonia [22–24] according
to reaction scheme (2).
LiNH2 ðsÞ þ H2 ðgÞ $ LiHðsÞ þ NH3 ðgÞ
(2)
This reaction reveals a rare example where hydrogen is converted, 8.1 wt%, and heat is absorbed, DH = 50 9 kJ/mol [25–27].
The reaction appears to be driven by larger entropy of the ammonia molecule than the hydrogen.
In reaction scheme (1), only the amide–imide reaction is reversible at moderate conditions with a hydrogen storage capacity of
rm = 6.5 wt% and an enthalpy change of DHdec = 66 kJ/mol H2
[28]. Several other reversible nitrogen-based systems have been
investigated recently and are promising for future mobile applications, for example, Mg(NH2)2–2LiH with a low calculated decomposition temperature, Tdec < 90 8C (at p(H2) = 1 bar) as a result of a
low enthalpy change for the reaction, DHdec = 38.9 kJ/mol H2 [29–
31].
Metal borohydrides and a metal amide can also form RHC
systems as shown for LiBH4–2LiNH2 in reaction scheme (3) [32].
Initially, a new crystalline solid with composition Li3BN2H8,
forms by mechanochemical treatment, which is an intermediate
compound prior to the formation of a very stable decomposition
product, Li3BN2, illustrated in reaction scheme (3). The onset
temperature for hydrogen release decreases from 380 to 250 8C
for LiBH4 in the composite LiBH4–2LiNH2, and thermal
130
decomposition releases more than 10 wt% H2 in the temperature
range 250–350 8C, but, unfortunately, the system is irreversible.
LiBH4 ðsÞ þ 2LiNH2 ðsÞ ! Li3 BN2 H8 ðsÞ
! Li3 BN2 ðsÞ þ 4H2 ðgÞ
(3)
In contrast, the analog lithium alanate amide system, LiAlH4–
2LiNH2, releases 2 equiv. H2 already during mechanochemical
treatment forming an amorphous mixture with the overall composition Li3AlN2H4, see reaction scheme (4a) [33,34]. Lithium
aluminum nitride, Li3AlN2, absorbs more than 5 wt% H2 forming
2LiH, LiNH2 and AlN, reaction (4b), which is another new RHC
system.
LiAlH4 ðsÞ þ 2LiNH2 ðsÞ ! ½Li3 AlN2 H4 ðsÞ þ 2H2 ðgÞ
! Li3 AlN2 ðgÞ þ 4H2 ðgÞ
(4a)
2LiHðsÞ þ LiNH2 ðsÞ þ AlNðsÞ $ Li3 AlN2 ðsÞ þ 2H2 ðgÞ
(4b)
The most successful reactive hydride composite was discovered
independently by the research groups of Vajo, HRL Laboratories,
California, USA and Dornheim, Klassen, Bormann and co-workers
at HZG, Hamburg, Germany, namely LiBH4–MgH2 [35,36]. The
great advantage is that the endothermic dehydrogenation of the
two hydrides in the composite 2LiBH4–MgH2 is followed by the
exothermic formation of MgB2, see reaction (5). The total reaction
enthalpy is thereby lowered to a calculated value of, DHdec 46 kJ/
(mol H2) corresponding to a calculated decomposition temperature of T 169 8C (at p(H2) = 1 bar) [35,37,38]. The hydrogen
absorption is facilitated from the MgB2–LiH composite and occurs
at p(H2) = 50 bar and T < 300 8C. These conditions are substantially more favorable than those of LiBH4 and are considered a
breakthrough in utilizing borohydrides for reversible hydrogen
storage. The full reversibility of the LiBH4–MgH2 system is only
obtained when the decomposition occurs in a hydrogen back
pressure of p(H2) 1–5 bar, which facilitates the formation of
MgB2 possibly owing to suppression of the individual decomposition of LiBH4 [38–41]. In fact, hydrogen release and uptake is a twostep reaction as shown in reaction scheme (5).
2LiBH4 ðsÞ þ MgH2 ðsÞ $ 2LiBH4 ðsÞ þ MgðsÞ þ H2 ðgÞ
$ 2LiHðsÞ þ MgB2 ðsÞ þ 4H2 ðgÞ
(5)
A similar reaction occurs for a magnesium-rich system,
0.3LiBH4–MgH2, during decomposition in p(H2) > 1 bar, that is,
formation of MgB2, whereas a- and b-alloys of Li1xMgx are formed
under a dynamic vacuum [42–44]. A number of other promising
reactive hydride composites have also been described, for example,
NaBH4–MgH2 and Ca(BH4)2–MgH2 [36,45–50]. Additionally, some
multicomponent systems have been developed such as LiBH4–
MgH2–LaH3 and Ca(BH4)2–LiBH4–MgH2. They benefit from high
cyclic stability and high hydrogen capacity [51,52].
Dihydrogen-bonding – a new approach for hydrogen
elimination
The strength and directionality of dihydrogen bonds, Hd+ –dH,
appear to be comparable to conventional hydrogen bonds. The
three intermolecular O–Hd+ –dH–B dihydrogen bonds in the
compound NaBH42H2O are in the range 1.77–1.95 Å and shorter
than H H distances within the BH4– anion, which are ca. 2.0 Å
[53]. Indeed, it suggests that the dihydrogen bonds may facilitate
hydrogen elimination during thermolysis at moderate temperatures. Metal borohydrides are often hygroscopic and in some cases
new crystalline compounds are formed by the absorption of water.
NaBH42H2O decomposes upon heating at 40 8C to NaBH4 and
2H2O, which at T > 40 8C slowly react to release hydrogen. Thus,
the hydrate NaBH42H2O does not directly release hydrogen, but
decomposes into anhydrous NaBH4 and water [53]. Similarly,
lithium borohydride exposed to air releases hydrogen at 65 8C
possibly due to a reaction between LiBH4 and H2O [54].
Metal borohydride ammoniates, M–BH4–NH3
Ammonia, NH3, is catalytically split to N2 and H2 and is a candidate for on-board hydrogen storage as a result of its high hydrogen
content (17.3 wt%) and the ability to store 30% more energy pr.
volume than liquid hydrogen [55]. However, due to the toxicity of
NH3 there are substantial safety issues that hamper widespread
utilization. Ammonia reacts with metal borohydrides by coordination to the metal and by formation of dihydrogen bonds to
BH4–. Metal borohydride ammoniates, M(BH4)mnNH3, were discovered in the 1950s and have recently attracted significant attention
as potential hydrogen storage materials, mainly for three reasons.
First, metal borohydride ammoniates often have high hydrogen
capacities and significantly lower dehydrogenation temperatures
compared to the metal borohydride owing to the dihydrogen
elimination of hydrogen. Destabilization is observed for borohydrides with low electronegativity. For example, Co-catalyzed
Li(NH3)4/3BH4 (i.e. 2/3Li(NH3)BH4 and 1/3Li(NH3)2BH4 with
equivalent protic and hydridic hydrogen atoms), releases ca.
17.8 wt% of H2 in a closed system in the temperature range
135–250 8C, in contrast to LiBH4 which release H2 at T > 380 8C,
8C, see Fig. 1 [56]. In fact, this is a solid state-gas reaction between
RESEARCH
LiBH4 and NH3. Secondly, unstable metal borohydrides with a
high electronegativity are stabilized by NH3 as demonstrated for
several metal borohydrides, such as Zn(BH4)2nNH3 and
Al(BH4)3nNH3 [57,58]. Ammonia always coordinates directly to
the metal and may prevent formation of neutral volatile molecular
borohydrides or reduction of the metal. Al(BH4)3 is among the
borohydrides with highest capacity (16.9 wt%), but is unstable and
volatile (Tbp 44 8C). However, Al(BH4)36NH3 is stable and
releases 11.8 wt% of H2 (purity 95%) with Tmax at 168 8C [59].
Thirdly, the composition of the released gas depends on the ratio
between NH3 and BH4 coordinated to the metal, excess of NH3
provides increased tendency to release ammonia, that is, ammonia
release from M(BH4)mnNH3 for n/m > 1 [60]. For instance, NH3 is
released from Mg(BH4)26NH3, while mainly H2 is released from
Mg(BH4)22NH3, see Fig. 2 [61]. Metals with low electronegativity
tend to release NH3 upon heating in open systems (p(NH3) 0),
but H2 in closed systems [56,62]. Metals with higher electronegativity coordinate more strongly to NH3 giving rise to a collapse of
the structure and release of H2 by dihydrogen elimination in the
temperature range 100 to 200 8C. This may prevent the release
of diborane from the more unstable metal borohydrides. A correlation between decomposition temperature and electronegativity
of the metal coordinating to NH3 and BH4 is observed in Fig. 1
represented by a dotted line.
In addition to the monometallic borohydride ammoniates, a
few bimetallic borohydride ammoniates have been synthesized
[60,63–65]. The first example, Li2Al(BH4)56NH3, has a remarkable
structure consisting of ordered [Al(NH3)6]3+ ammine complexes
and [Li2(BH4)5]3 complex anions and reveal attractive decomposition properties [60].
Ammonia is detrimental for low temperature fuel cells, and the
NH3/BH4 ratio (n/m) requires tailoring to avoid ammonia release.
Metal borohydrides readily react with ammonia to obtain
M(BH4)mnNH3. A series of calcium borohydride ammoniates,
Ca(BH4)2nNH3 (n = 1, 2, 4, 6), can be obtained by a sequential
heating procedure [62,66]. Some metal borohydride ammoniates
with low n/m ratio have successfully been prepared by metathesis
reactions between metal chloride ammoniates and lithium borohydride [67]. The amount of NH3 is adjusted by partial release of
NH3 from the metal chloride prior to the synthesis. The mechanochemical approach facilitates formation of Zn(BH4)22NH3, while
Zn(BH4)24NH3 is obtained by solvent-based methods [57]. This
method introduces significant amounts of LiCl in the final product. Thus, development of new synthesis routes providing halide
free materials with specific n/m ratios is important.
Metal borohydride amides, M–BH4–NH2
FIGURE 1
Experimentally observed decomposition temperatures, Tdec, for selected
metal borohydrides and metal borohydride ammoniates plotted as a
function of the electronegativity, xp, of the metal. Metal borohydrides with
low electronegativity are destabilized by ammoniate formation while those
with higher electronegativity are stabilized. The dashed line indicates a
correlation between decomposition temperatures and electronegativity for
metal borohydrides ammoniates and a new approach for rational design of
materials properties.
An alternative approach to combine partially positive H atoms to
hydridic hydrogen in metal borohydrides is by using metal amides,
M(NH2)n. Numerous combinations between alkali and alkaline
earth metal amides and borohydrides have been investigated, for
example, Mg(BH4)2–LiNH2 [68], Ca(BH4)2–Mg(NH2)2 [69], and
LiBH4–Mg(NH2)2 [70]. As discussed above, LiBH4–2LiNH2 is an
example of a reactive hydride composite with dihydrogen bonding
[32]. Changing the reactant ratio LiBH4–LiNH2 to (1:1) or (1:3)
gives rise to compounds with different chemical compositions,
Li2(BH4)(NH2) or Li4(BH4)(NH2)3, which are also good lithium ion
conductors [71]. The detailed crystal structures are known for
131
RESEARCH: Review
Materials Today Volume 17, Number 3 April 2014
RESEARCH
Materials Today Volume 17, Number 3 April 2014
RESEARCH: Review
FIGURE 2
The crystal structures of Mg(BH4)26NH3 (top) and Mg(BH4)22NH3. NH3 (N green, H light gray) coordinates to magnesium (orange) in the crystal structures,
while BH4 (blue tetrahedra) coordinates either to the metal or acts as a counter ion in the solid state [61]. For clarity, hydrogen atoms in the ammonia
molecules are not shown in Mg(BH4)26NH3.
Li4(BH4)(NH2)3 and Li2(BH4)(NH2) revealing that the borohydride
and amide groups remain [72–74].
To the best of our knowledge, the LiBH4–LiNH2 system is unique
compared to the other metal borohydride amide systems in the
sense that new quaternary structures are readily formed by
mechanochemical treatment. In all other systems a physical mixture is obtained. The physical mixtures typically react during
thermal treatment and decompose at lower temperatures compared to the individual components, but without forming borohydride-amide complexes. However, recently Mg(BH4)(NH2) was
prepared by combined mechanochemical and thermal treatment
of Mg(BH4)2–Mg(NH2)2 [75].
Ammonia borane and derivatives
Ammonia borane, NH3BH3 (AB) has attracted significant attention
owing to its extreme hydrogen capacities of rm = 19.6 wt% H2,
rV = 146 g H2/L, and air stability [76]. Solid crystalline ammonia
borane has an intermolecular dihydrogen bond network and is not
hygroscopic unlike borohydrides. This compound, NH3BH3 releases
one equivalent of hydrogen in each of the three decomposition
steps forming polyaminoborane, [NH2BH2]n (90–120 8C), polyiminoborane, [NHBH]n (120–200 8C) and finally boron nitride, BN
(>500 8C) [76]. However, the hydrogen release is accompanied by
toxic by-products, such as ammonia (NH3), diborane (B2H6) and
borazine (N3B3H3). Furthermore, the decomposition is exothermic
(DHdec = 21 kJ/mol H2), hence non-reversible [77]. However, significantly improved properties of NH3BH3 were obtained by infiltration in ordered mesoporous silica facilitating enhanced kinetics
(faster hydrogen release) at lower temperatures with reduced borazine emission and improved thermodynamics, DHdec = 1 kJ/mol
H2 [78]. Additionally, the work by Autrey and co-workers [78]
132
initiated the focus on nanoconfinement as a tool to improve
kinetics and possibly thermodynamics of hydrogen storage materials [79]. More recently, nano-sized Co and Ni additives in ammonia
borane were observed to improve the kinetics and suppress borazine
emission and foaming [80].
An even more hydrogen-rich compound can be prepared: ammonium borohydride, NH4BH4. It has the highest hydrogen content of
solid-state materials reported to date (rm = 24.5 wt% H2, rv = 151 g
H2/L) and releases 3 equivalents of hydrogen (18 wt% H2) in three
distinct exothermic steps at T < 160 8C [81]. However, NH4BH4
slowly decomposes at RT with a half-life of 6 h to a diammoniate
of diborane (NH3)2BH2(BH4) (DADB) and hydrogen. For long-term
storage NH4BH4 must be kept at T < 40 8C. Recently, nanoconfined NH4BH4 in mesoporous silica was investigated and appears to
be destabilized and more rapidly decomposes to DADB [82].
The complicated synthesis methods have hampered the
detailed investigation of (NH3)2BH2(BH4). However, recently a
mechanochemical reaction between NH4BH4 and NH3BH3 was
discovered [83], see scheme (6).
NH4 BH4 ðsÞ þ NH3 BH3 ðsÞ ! ½ðNH3 Þ2 BH2 ½BH4 ðsÞ þ H2 ðgÞ
(6)
DADB decomposes in reaction steps similar to AB, but with a
slightly lower onset temperature, faster kinetics and no significant
induction period prior to hydrogen release, possibly due to DADB
known to be an intermediate in the decomposition of AB [84].
Metal amidoboranes, M(NH2BH3)n, are synthesized by reacting a
metal hydride with ammonia borane using either mechanochemistry or solvent-based methods [85], see reaction scheme (7).
MHn ðsÞ þ nNH3 BH3 ðsÞ ! MðNH2 BH3 Þn ðsÞ þ nH2 ðgÞ
(7)
Materials Today Volume 17, Number 3 April 2014
RESEARCH
LiNH2 BH3 ðsÞ ! LiNBHðsÞ þ 2H2 ðgÞ
(8)
Metal amidoborane ammoniates, M(NH2BH3)nxNH3 are known
for M = Mg, Ca [94,95]. These compounds tend to release NH3
below 100 8C in an open system (endothermic reaction) and H2 in
a closed system (exothermic). Recently, hydrogen release was
observed from an endothermic reaction from composites of
Mg(NH2BH3)22NH3 and NaH/KH [96].
In 2010, the first metal borohydride–ammonia borane complexes,
M–BH4–NH3BH3 were reported, that is, Li2(BH4)2NH3BH3 and
Ca(BH4)2(NH3BH3)2 [97]. Since then, LiBH4(NH3BH3) [98], and
Mg(BH4)2(NH3BH3)2 [99,100] have also been reported, and they
are all prepared by mechanochemical treatment of AB and
M(BH4)n (M = Li, Mg, Ca). In contrast, MBH4–NH3BH3 (M = Na,
K, Cs, Rb) do not form new compounds during mechanochemical
treatment [99,101]. This class of materials have high hydrogen
capacities and low decomposition temperatures, for example,
Mg(BH4)2(NH3BH3)2 (rm = 17.4 wt% H2, rV = 137 g H2/L) melts
at 48 8C and has an onset temperature for hydrogen release at
75 8C [99]. However, the thermal decomposition of these compounds still involves the release of diborane and borazine similar
to NH3BH3 reflecting a weak interaction between the borohydride
groups and ammonia borane. This is explained by the crystal
structures of M(BH4)n(NH3BH3)x, where AB keeps its molecular
form [97]. The crystal structure of Mg(BH4)2(NH3BH3)2 is shown in
Fig. 3. Both borohydride groups and AB act as terminal ligands,
and molecular complexes are linked in the crystal structure via
dihydrogen bonds of N–Hd+ –dH–B (<2 Å).
Hydrazine and hydrazine borane
Hydrazine, N2H4 (12.5 wt% H2) decomposes via two competing
reactions forming N2, H2 and NH3. Recently a new class of hydrogen storage materials, borohydride hydrazinates, was successfully
synthesized, for example, LiBH4NH2NH2 and LiBH42NH2NH2
[102]. Approximately 13.0 wt% H2 is released from LiBH4NH2NH2
2NH2 at 140 8C in the presence of Fe–B catalysts. However, this
again leads to the formation of the stable compounds Li3BN2 and
BN according to reaction scheme (9).
3ðLiBH4 N2 H4 ÞðsÞ ! Li3 BN2 ðsÞ þ 2BNðsÞ
þ N2 ðgÞ þ 12H2 ðsÞ
RESEARCH: Review
Since 2007, a series of metal amidoboranes, M(NH2BH3)n
(M = Li, Na, K, Ca, Sr), have been synthesized according to reaction
scheme (7) and structurally investigated [86–89]. In contrast,
Y(NH2BH3)3 is synthesized by a metathesis reaction between
MNH2BH3 and YX3 (M = Li, Na; X = Cl, F) [90]. The magnesium
analog, Mg(NH2BH3)2, was previously considered unstable, but
was recently synthesized [91]. Metathesis reaction between FeCl3
and LiNH2BH3 in THF was unsuccessful, but LiCl and polymeric
[Fe(HN = BH)3]n was formed together with the release of 1.5 equiv.
H2 in THF at RT [92]. In general, metal amidoboranes have high
hydrogen content, good kinetics and low decomposition temperatures. LiNH2BH3 decomposes in the temperature range from 75 to
95 8C and releases 10.9 wt% H2 according to scheme (8) [86].
Formation of metal amidoboranes is considered an approach to
prevent release of borazine from ammonia borane [93].
FIGURE 3
Crystal structure of Mg(BH4)2(NH3BH3)2. Molecular complexes of
[Mg(BH4)2(NH3BH3)2] are connected by dihydrogen bonds (dotted lines).
Mg, N, B and H are represented by orange, green, blue and light gray
spheres and the [BH4] complexes as blue tetrahedra [99].
borane (N2H4BH3, HB). HB (15.4 wt% H2) melts at 61 8C at which
point the decomposition initiates yielding NH2NH2, (NHBH2)2, H2
and NH3. Interestingly, mechanochemical treatment of
LiH–HB (1:1) and (1:3) provided the first metal hydrazinoborane,
LiN2H3BH3, and its hydrazine borane adduct LiN2H3
BH32N2H4BH3 [103]. The metal hydrazinoboranes exhibit dramatically improved dehydrogenation properties compared to hydrazine borane (N2H4BH3) with nearly complete dehydrogenation in
the temperature range 50–225 8C releasing high purity hydrogen.
However, the decomposition for both compounds is exothermic,
hence non-reversible.
Reversibility of B–N based hydrides
Generally, the B–N compounds discussed in this review paper
show high hydrogen storage capacities both gravimetrically and
volumetrically and often release hydrogen at low temperatures see
Table 1. However, they all suffer from limited reversibility due to
the formation of stable boron nitrides in the decomposed residue.
Therefore, further research in kinetics and thermodynamics
(9)
Furthermore, the decomposition product of AB, polyiminoborane reacts with hydrazine in THF solution to form hydrazine
FIGURE 4
Reaction mechanism for ammonia borane, NH3BH3, regeneration from
polyiminoborane, ‘BNH’ using hydrazine, N2H4 (modified from Ref. [108]).
133
RESEARCH
Materials Today Volume 17, Number 3 April 2014
TABLE 1
Selected solid boron- and nitrogen-based hydrides with high gravimetric (rm) and volumetric (rV ) hydrogen densities and often with
low onset temperatures for hydrogen release (Tonset). Unfortunately, the released hydrogen gas may contain varying amounts of
impurities.
RESEARCH: Review
Compound
rm (wt%)
rV (g H2/L)
Tonset (8C)
Impurities
Ref.
LiBH4
18.5
122.5
380
Not observed
[104]
a-Mg(BH4)2
14.9
117
250
Not observed
[105]
Mg(BH4)22NH3
16.0
130
150
NH3
[61]
Li2Al(BH4)56NH3
17.6
151
75
NH3 (>99% H2)
[60]
Al(BH4)36NH3
17.4
150
130
NH3 (95% H2)
[59]
Zn(BH4)22NH3
10.8
135
90
Not observed
[57]
NH4BH4
24.5
151
85
B2H6, B3N3H6, NH3
[81]
NH3BH3
19.4
144
108
B2H6, B3N3H6, NH3
[76]
LiNH2BH3
10.9
92
85
NH3
[86]
Mg(BH4)2(NH3BH3)2
17.4
137
75
B2H6, B3N3H6, NH3
[99]
Li2(BH4)2NH3BH3
19.0
138
105
B2H6, B3N3H6, NH3
[97]
related to hydrogen uptake is needed and further investigation of
partly decomposed B–N–H materials may be fruitful.
Hydrogen uptake in fully decomposed Mg(BH4)2 is possible, but
requires extreme conditions: 500 8C and p(H2) = 950 bar [106,107].
However, partial dehydrogenation of Mg(BH4)2 at lower temperatures (250 8C) forms Mg(B3H8)2, which is more readily rehydrogenated (250 8C, p(H2) = 120 bar, 48 h). A similar approach may be
successful for B–N–H based compounds. Recently, the regeneration of ammonia borane is reported to take place from polyiminoborane by reacting with hydrazine in liquid ammonia at 40 8C
within 24 h [108]. Fig. 4 illustrates the ideal cycle for reversible
hydrogen storage using NH3BH3. Importantly, this cycle is closed
and the generation of AB from hydrazine takes place in one step
and does not involve any noble metal catalysts. This discovery is
among the most important breakthroughs for possible utilization
of ammonia borane for hydrogen storage and at the same time
reveals new classes of materials based on ammonia borane and
hydrazine.
Conclusion
This review illustrates the extreme diversity in the fascinating
chemistry of hydrogen, regarding the variety of chemical bonds
and compounds that can be created. The kinetic and thermodynamic properties and hydrogen storage densities can be tailored,
which reveal new perspectives for the development of solid-state
hydrogen storage materials. Dihydrogen bonding provides hydrogen elimination at moderate temperatures but reformation of this
type of bond is difficult. A similar drawback is observed for
ammoniates and amides of metal borohydrides, which otherwise
provide extreme hydrogen densities and low decomposition temperatures. Fortunately, the recent discovery of ammonia borane
regeneration using hydrazine reveals that further research in
partial dehydrogenation may provide new reversible reaction routes
for hydrogen release and uptake at moderate conditions. An overwhelming variety of novel boron and nitrogen based materials have
been discovered over the past few years, which provide new
approaches for the rational design of materials with tailored properties and new hope for the discovery of novel types of hydrogen
134
storage materials. Further research within boron and nitrogen
based hydrides may significantly support the implementation of
hydrogen as a future energy carrier, for applications including mobile
devices, in a sustainable future for humanity.
Acknowledgements
The work was supported by the Danish Council for Strategic
Research via the research project HyFillFast, the Danish National
Research Foundation, Center for Materials Crystallography
(DNRF93), and by European Community FP7/2007–2013, FCH JU
under grant agreement no. 303428 – Bor4Store and by the COST
Action MP1103 ‘Nanostructured materials for solid-state hydrogen
storage’. We are grateful to the Carlsberg Foundation.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
D.J.C. MacKay, Sustainable Energy – Without the Hot Air, UIT, Cambridge, 2009.
J.A. Ritter, et al. Mater. Today 6 (9) (2003) 18.
A. Züttel, Mater. Today 6 (9) (2003) 24.
A. Züttel, A. Borgschulte, L. Schlapbach, Hydrogen as a Future Energy Carrier,
Wiley-VCH Verlag GmbH & Co., Weinheim, 2008.
G. Walker, Solid-state Hydrogen Storage: Materials and Chemistry, Woodhead
Publishing, 2008.
L. Schlapbach, A. Züttel, Nature 414 (6861) (2001) 353.
W.I.F. David, Faraday Discuss. 151 (2011) 399.
M. Hirscher, Handbook of Hydrogen Storage, Wiley-VCH Verlag GmbH & Co,
Germany, Weinheim, 2010.
U. Eberle, et al. Angew. Chem. Int. Ed. 48 (36) (2009) 6608.
S.I. Orimo, et al. Chem. Rev. 107 (10) (2007) 4111.
L.H. Rude, et al. Phys. Status Solidi 208 (2011) 1754.
J. Lu, et al. Inorg. Chem. 45 (21) (2006) 8749.
L.H. Rude, et al. J. Phys. Chem. C 116 (38) (2012) 20239.
D.R. Lide, CRC Handbook of Chemistry and Physics, 88th ed., CRC Press, 2007.
M.-A. Richard, et al. Angew. Chem. Int. Ed. 55 (11) (2009) 2985.
M. Hirscher, B. Panella, Scr. Mater. 56 (10) (2007) 809.
B. Panella, M. Hirscher, Adv. Mater. 17 (5) (2005) 538.
M.B. Ley, et al. Mater. Today 17 (3) (2014) 121.
W. Grochala, P.P. Edwards, Chem. Rev. 104 (3) (2004) 1283.
M. Dornheim, et al. Adv. Eng. Mater. 8 (5) (2006) 377.
P. Chen, et al. Nature 420 (6913) (2002) 302.
W.I.F. David, et al. J. Am. Chem. Soc. 129 (6) (2007) 1594.
T. Ichikawa, et al. J. Phys. Chem. B 108 (23) (2004) 7887.
Y.H. Hu, E. Ruckenstein, J. Phys. Chem. A 107 (46) (2003) 9737.
Y. Kojima, et al. J. Mater. Res. 24 (07) (2009) 2185.
S. Hino, et al. J. Appl. Phys. 105 (2) (2009) 023527.
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
H. Yamamoto, et al. Int. J. Hydrogen Energy 34 (24) (2009) 9760.
Y. Kojima, Y. Kawai, J. Alloys Compd. 395 (1/2) (2005) 236.
K.C. Kim, et al. Phys. Chem. Chem. Phys. 13 (15) (2011) 7218.
Z. Xiong, et al. Adv. Mater. 16 (17) (2004) 1522.
W.F. Luo, J. Alloys Compd. 385 (1–2) (2004) 316.
F.E. Pinkerton, et al. J. Phys. Chem. B 109 (1) (2005) 6.
Xiong, et al. Adv. Funct. Mater. 17 (7) (2007) 1137.
J. Huot, et al. Prog. Mater. Sci. 58 (1) (2013) 30.
J.J. Vajo, et al. J. Phys. Chem. B 109 (9) (2005) 3719.
G. Barkhordarian, et al. J. Alloys Compd. 440 (1/2) (2007) L18.
Y.W. Cho, et al. Calphad 30 (1) (2006) 65.
U. Bösenberg, et al. Acta Mater. 55 (11) (2007) 3951.
U. Bösenberg, et al. Acta Mater. 58 (2010) 3381.
U. Bösenberg, et al. J. Phys. Chem. C 114 (35) (2010) 15212.
J.H. Shim, et al. J. Phys. Chem. Lett. 1 (1) (2009) 59.
T.E.C. Price, et al. J. Am. Chem. Soc. 133 (34) (2011) 13534.
O. Friedrichs, et al. Phys. Chem. Chem. Phys. 11 (10) (2009) 1515.
X.B. Yu, et al. Chem. Commun. 37 (2006) 3906.
G. Barkhordarian, et al. J. Phys. Chem. C 112 (7) (2008) 2743.
T.K. Nielsen, et al. ACS Nano 4 (7) (2010) 3903.
R. Gosalawit-Utke, et al. J. Phys. Chem. C 114 (22) (2010) 10291.
R. Gosalawit-Utke, et al. Int. J. Hydrogen Energy 38 (8) (2013) 3275.
K.S. Alcantara, et al. J. Phys. Chem. C 116 (12) (2012) 7207.
C. Pistidda, et al. J. Phys. Chem. C 114 (49) (2010) 21816.
M. Gao, et al. J. Mater. Chem. A 1 (39) (2013) 12285.
Y. Zhou, et al. J. Phys. Chem. C 116 (1) (2012) 1588.
Y. Filinchuk, H. Hagemann, Eur. J. Inorg. Chem. 20 (2008) 3127.
L. Mosegaard, et al. J. Alloys Compd. 446–447 (2007) 301.
P.J. Feibelman, et al. Phys. Today 58 (6) (2005) 13.
X. Zheng, et al. Energy Environ. Sci. 4 (9) (2011) 3593.
Q. Gu, et al. Energy Environ. Sci. 5 (6) (2012) 7590.
Y. Guo, et al. Angew. Chem. Int. Ed. 50 (5) (2011) 1087.
Y. Guo, et al. Chem. Commun. 48 (37) (2012) 4408.
Y. Guo, et al. J. Am. Chem. Soc. 133 (13) (2011) 4690.
G. Soloveichik, et al. Inorg. Chem. 47 (10) (2008) 4290.
H. Chu, et al. Chem. Mater. 22 (21) (2010) 6021.
G. Xia, et al. J. Mater. Chem. 22 (15) (2012) 7300.
W. Sun, et al. Chem. Eur. J. 18 (22) (2012) 6825.
F. Yuan, et al. Int. J. Hydrogen Energy 38 (13) (2013) 5322.
Z. Tang, et al. J. Mater. Chem. 22 (12) (2012) 5312.
F. Yuan, et al. Chem. Mater. 24 (17) (2012) 3370.
RESEARCH
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
[100]
[101]
[102]
[103]
[104]
[105]
[106]
[107]
[108]
X.B. Yu, et al. J. Phys. Chem. C 114 (10) (2010) 4733.
H. Chu, et al. Dalton Trans. 39 (44) (2010) 10585.
X.Y. Chen, et al. J. Phys. Chem. C 114 (41) (2010) 17947.
M. Matsuo, et al. J. Am. Chem. Soc. 131 (45) (2009) 16389.
Y.E. Filinchuk, et al. Inorg. Chem. 45 (4) (2006) 1433.
P.A. Chater, et al. Chem. Commun. 45 (2007) 4770.
H. Wu, et al. Chem. Mater. 20 (4) (2008) 1245.
T. Noritake, et al. Int. J. Hydrogen Energy 38 (16) (2013) 6730.
A. Staubitz, et al. Chem. Rev. 110 (7) (2010) 4079.
G. Wolf, et al. Thermochim. Acta 343 (1/2) (2000) 19.
A. Gutowska, et al. Angew. Chem. Int. Ed. 44 (23) (2005) 3578.
T.K. Nielsen, et al. Nanoscale 3 (2011) 2086.
T. He, et al. Chem. Mater. 21 (11) (2009) 2315.
A. Karkamkar, et al. Chem. Mater. 21 (19) (2009) 4356.
T.K. Nielsen, et al. Dalton Trans. 43 (2013) 680.
M. Bowden, et al. Chem. Commun. 46 (45) (2010) 8564.
Z. Fang, et al. Phys. Chem. Chem. Phys. 13 (16) (2011) 7508.
Y.S. Chua, et al. Chem. Commun. 47 (2011) 5116–5129.
Z. Xiong, et al. Nat. Mater. 7 (2) (2008) 138.
H.V.K. Diyabalanage, et al. J. Am. Chem. Soc. 132 (34) (2010) 11836.
H.V.K. Diyabalanage, et al. Angew. Chem. Int. Ed. 119 (47) (2007) 9153.
Q. Zhang, et al. J. Phys. Chem. C 114 (3) (2010) 1709.
R.V. Genova, et al. J. Alloys Compd. 499 (2) (2010) 144.
J. Luo, et al. Energy Environ. Sci. 6 (3) (2013) 1018.
T. He, et al. J. Mater. Chem. 22 (15) (2012) 7478.
H. Wu, et al. J. Am. Chem. Soc. 130 (44) (2008) 14834.
Y.S. Chua, et al. Chem. Commun. 46 (31) (2010) 5752.
Y.S. Chua, et al. Chem. Mater. 21 (20) (2009) 4899.
Y.S. Chua, et al. Chem. Mater. 24 (18) (2012) 3574.
H. Wu, et al. J. Mater. Chem. 20 (31) (2010) 6550.
J. Luo, et al. Int. J. Hydrogen Energy 37 (14) (2012) 10750.
L.H. Jepsen et al., 2013 (submitted for publication).
X. Chen, et al. Dalton Trans. 42 (40) (2013) 14365.
L.H. Jepsen, et al. J. Alloys Compd. 580 (2013) S287–S291.
T. He, et al. Energy Environ. Sci. 5 (2) (2012) 5686.
H. Wu, et al. Energy Environ. Sci. 5 (6) (2012) 7531.
J.P. Soulie, et al. J. Alloys Compd. 346 (1/2) (2002) 200.
Y. Filinchuk, et al. Chem. Mater. 21 (5) (2009) 925.
M. Chong, et al. Chem. Commun. 47 (4) (2011) 1330.
C. Pistidda, et al. J. Alloys Compd. 508 (1) (2010) 212.
A.D. Sutton, et al. Science 331 (6023) (2011) 1426.
RESEARCH: Review
Materials Today Volume 17, Number 3 April 2014
135